Addressing risk and uncertainty in long-term PPP contracts

June 22, 2018, Paris

Alexander Galetovic, Universidad de los Andes

Working Group Papers , International Transport Forum, 2018

1. Makovšek, D., “Mobilising Private Investment in

Infrastructure: Investment De-Risking and Uncertainty”

2. Vasallo, J. M., “Public-Private Partnerships in Transport:

Unbundling Prices from User Charges”

3. Engel, E., R. Fischer and A. Galetovic, “Dealing with the

Obsolescence of Transport Infrastructure in Public-Private

Partnerships”

Idea 1: private participation in transport infrastructure is modest

3

Total and PPP investment in road and rail infrastructure in OECD7* countries

(1995-2014, in USD million 2005 prices)

4

Source: Makovsek (2018)

0

20 000

40 000

60 000

80 000

100 000

120 000

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

Private investment in rail and road in infrastructure Total investment in rail and road infrastructure

2005 prices, US$ million

Annual world infrastructure spending, (2008-10, USD billion) 5

Total Private

(1)

Públic+ private

(2)

PPP

(3)

Project finance

(4)

Corporate

Transport 1,040 [45─75] ─ nd

Airports 80

Seaports 110

Rail 400

Roads 450

Social infrastructure 490 [12─20] ─ na

Water & sanitation 160

[3─5]

─ na

Oil & gas 200 na na

Electricity 810 [140─160] na

Telecommunications 300 [42─48] na

Total 3,000

Total private 1,000 [60─100] [180─220] [680─760]

World GDP (2010) 63.000

Idea 2: PPPs have worked well in seaports and airports; less so in roads and rail

6

PPP performance in transport infrastructure

• Seaports and airports (≈ 20 percent of total transport infra spending)

– Drewry database: 252 landlord container ports (2014)

– PPIAFF database: 141 private/concessioned airports (2014)

– Private investment accomodated massive increases in capacity & trade

• Rail, roads, tunnels and bridges (≈ 80 percent of total transport infra spending)

– Small participation in general, and concentrated in a few countries

– Contracts are often renegotiated

7

Idea 3: risk and insufficient funding are different problems

8

–Funding: who will pay for the project

(users/tolls; taxpayers/budget; a combination

→ see Vasallo)

–Risk: unpredictable variation in total project

value (Revenues ─ Costs): you don’t know

which outcome will realize

9

Idea 4: the key question about a risk is: can some party do something about it?

10

The answers

• Yes (endogenous risks)

− Make good outcomes more likely, and bad oucomes less likely

− De-risking: invest/spend to clarify what the risk is about or reduce the magnitude of the unpredictable variation

− The questions: (i) who should be responsible for the risk; (ii) what do you get in exchange for the risk transfer

• No (exogenous risks):

– Who is best suited to bear/spread the risk?

– The party that bears exogenous risk “sells” insurance

– The question: who should sell insurance to whom?

11

Idea 5: thoughtful risk allocation is a de-risking strategy by itself

12

An example: demand risk in ports and roads

• Seaports: quality & speed of service can affect

demand for the port dramatically → substantial

part of demand risk is endogenous → let the PPP

bear demand risk

• Roads: largely exogenous once road is available

→ should the government “buy” or “sell”

insurance?

13

Fixed term PPPs create demand risk ($600)

14

Year 1 2 3 4 5 6 7 8 9 10 11 12

High demand (p = 0.5)

100 100 100 100 100 100 100 100 100 100 100 100

Low demand (p = 0.5)

50 50 50 50 50 50 50 50 50 50 50 50

Expected traffic 75 75 75 75 75 75 75 75 75 75 75 75

Roads and demand risk

• Hard to think that PPP investors are in the insurance

business (SPV + capital market?)

• Government can spread risk among taxpayers

• Private infrastructure (terminals, pipelines) use take-

or-pay contracts

• Lot’s of renegotiations because demand was too low

15

An availability contract ($600)

16

Year 1 2 3 4 5 6 7 8 9 10 11 12

High demand (p = 0.5)

100 100 100 100 100 100 100 100 100 100 100 100

Low demand (p = 0.5)

50 50 50 50 50 50 50 50 50 50 50 50

Expected traffic 75 75 75 75 75 75 75 75 75 75 75 75

A Least-Present-Value-of-Revenues contract ($600)

17

Year 1 2 3 4 5 6 7 8 9 10 11 12

High demand (p = 0.5)

100 100 100 100 100 100 100 100 100 100 100 100

Low demand (p = 0.5)

50 50 50 50 50 50 50 50 50 50 50 50

Expected traffic 75 75 75 75 75 75 75 75 75 75 75 75

Idea 6: the government has more and better options if the road PPP doesn’t bear demand risk

18

• PPP revenues and bids are tied to investment, not user demand, and fixed at the beginning of the concession

• A buyout call option is easier to value (Total − Extant Revenues); the government can retain flexibility without expropriating

–Option to revamp the infrastructure

–Option to bear obsolescence risk

• Separation of tolling and funding of PPPs: a road fund which pools availability-based contracts (Vasallo’s proposal)

19

Thank you!